1. And we begin again with the dispute over the homicide figures. Last week, Milenio stated that in January, 957 killings were registered in the whole country, a total of 1,939 in the first two months of the current six-year term. By Reforma’s count, January saw 769 homicides linked to organized crime, which places the total at 1,524 in the first two months of Peña Nieto’s government. For its part, La Jornada published some mysterious data, obtained from "federal employees that worked in the national security cabinet," according to which 1,758 people would have been killed in incidents related “to disputes between criminal groups” or in "confrontations."
2. Which of these numbers are correct? None. According to the Secretary General of National Public Safety (SESNSP), 1,569 previous confirmations of intentional homicide were registered in December. It is highly probable that January’s number will be similar, giving an approximate total of 3,200 intentional homicides in the first two months of the current administration (we will know the precise figure in a couple of weeks when SESNSP releases January’s information).
3. And how many of those 3,200 are linked to organized crime? Nobody knows and those who say they know, don’t. The independent murder counts (and the official count used until a few months ago) are constructed based on inferences: they use some characteristics of the incidents or of the victims (presumed use of a firearm, messages next to the corpse, etc.) in order to suppose that a homicide could have been connected with something vague called "organized crime." But they are no more than that – assumptions not validated by judicial investigations which could allow the motives behind the homicides to be explained.
4. Are 3,200 homicides committed during the first two months of the current government many or few? Well, it depends what you are comparing it against: compared to the 1,799 homicides that were committed in the first two months of the government of Felipe Calderon, they are a lot. Compared to the 3,354 that were committed in December 2011 and January 2012, they represent a slight reduction (of approximately -4.5 percent). Against the peak numbers (the 4,128 registered in the period May-June 2011), they signify a drop of 22 percent (although this comparison is somewhat deceptive – see the following point).
5. But did the violence increase in December compared to November, or not? Well yes, but those types of month-to-month comparisons, crude and without any adjustments, are absurd: a) there are calendar effects that complicate the comparisons (December has more days than November, for example, b) there are seasonal effects (there are more homicides in summer than in winter and in December, the count rises as an effect of year-end festivities), and c) a few situational occurrences can shift the totals without causing changes in the underlying trend.
6. And so, how are we doing? Is violence rising or falling in the current administration? We don’t know and we won’t know for a number of months. Until December, it was decreasing: in 2012, the number of intentional homicides dropped 8.5 percent compared to the year before, the first decrease at an annual rate since 2007. If one takes a mobile average of three months, in December it reached the lowest level since April 2010. Will this trend continue in 2013? No idea: I’ll tell you when I have the figures.
7. But in some regions the situation is getting worse, isn’t it? Yes, but in others it is getting better: there is Ciudad Juarez, but likewise, there is the case of Nuevo Leon: in the fourth trimester, the number of intentional homicides decreased 48 percent with respect to the same period in 2011. In general, there are more examples of improvement than deterioration: in 2012, the number of intentional homicides decreased in 20 of 32 federal entities.
8. But what about the killing sprees? Aren’t the case of Kombo Kolombia or the cascade of multiple homicides in the State of Mexico signs of an imminent explosion in the levels of violence? No: to use the terminology of Nate Silver, that is noise, not a signal. As spectacular and perturbing as they are, the massacres do not define the overall state of criminal violence in the country. Incidents involving ten or more victims are counted at a few tens each year: the grand majority of homicides have only one victim (or two at the most).
9. And what then? What do we do as more data appears? Demand the government to publish the information as quickly as possible: it is the least that can be asked of an administration that has made the reduction of violence the principal objective of its security policy.
Translated and reprinted with permission from *Alejandro Hope, of Plata o Plomo, a blog on the politics and economics of drugs and crime published by Animal Politico. Read the Spanish original here. Hope is also a member of InSight Crime's Board of Directors.